1
# Copyright (C) 2005 Canonical Ltd
3
# This program is free software; you can redistribute it and/or modify
4
# it under the terms of the GNU General Public License as published by
5
# the Free Software Foundation; either version 2 of the License, or
6
# (at your option) any later version.
8
# This program is distributed in the hope that it will be useful,
9
# but WITHOUT ANY WARRANTY; without even the implied warranty of
10
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
11
# GNU General Public License for more details.
13
# You should have received a copy of the GNU General Public License
14
# along with this program; if not, write to the Free Software
15
# Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
17
"""Tree classes, representing directory at point in time.
21
from cStringIO import StringIO
22
from warnings import warn
25
from bzrlib import delta, osutils
26
from bzrlib.decorators import needs_read_lock
27
from bzrlib.errors import BzrError, BzrCheckError
28
from bzrlib import errors
29
from bzrlib.inventory import Inventory
30
from bzrlib.inter import InterObject
31
from bzrlib.osutils import fingerprint_file
32
import bzrlib.revision
33
from bzrlib.trace import mutter, note
37
"""Abstract file tree.
39
There are several subclasses:
41
* `WorkingTree` exists as files on disk editable by the user.
43
* `RevisionTree` is a tree as recorded at some point in the past.
45
Trees contain an `Inventory` object, and also know how to retrieve
46
file texts mentioned in the inventory, either from a working
47
directory or from a store.
49
It is possible for trees to contain files that are not described
50
in their inventory or vice versa; for this use `filenames()`.
52
Trees can be compared, etc, regardless of whether they are working
53
trees or versioned trees.
56
def changes_from(self, other, want_unchanged=False, specific_files=None,
57
extra_trees=None, require_versioned=False):
58
"""Return a TreeDelta of the changes from other to this tree.
60
:param other: A tree to compare with.
61
:param specific_files: An optional list of file paths to restrict the
62
comparison to. When mapping filenames to ids, all matches in all
63
trees (including optional extra_trees) are used, and all children of
64
matched directories are included.
65
:param want_unchanged: An optional boolean requesting the inclusion of
66
unchanged entries in the result.
67
:param extra_trees: An optional list of additional trees to use when
68
mapping the contents of specific_files (paths) to file_ids.
69
:param require_versioned: An optional boolean (defaults to False). When
70
supplied and True all the 'specific_files' must be versioned, or
71
a PathsNotVersionedError will be thrown.
73
The comparison will be performed by an InterTree object looked up on
76
# Martin observes that Tree.changes_from returns a TreeDelta and this
77
# may confuse people, because the class name of the returned object is
78
# a synonym of the object referenced in the method name.
79
return InterTree.get(other, self).compare(
80
want_unchanged=want_unchanged,
81
specific_files=specific_files,
82
extra_trees=extra_trees,
83
require_versioned=require_versioned,
86
def iter_changes(self, from_tree, include_unchanged=False,
87
specific_file_ids=None):
88
intertree = InterTree.get(from_tree, self)
89
return intertree.iter_changes(from_tree, self, include_unchanged,
93
"""Get a list of the conflicts in the tree.
95
Each conflict is an instance of bzrlib.conflicts.Conflict.
99
def get_parent_ids(self):
100
"""Get the parent ids for this tree.
102
:return: a list of parent ids. [] is returned to indicate
103
a tree with no parents.
104
:raises: BzrError if the parents are not known.
106
raise NotImplementedError(self.get_parent_ids)
108
def has_filename(self, filename):
109
"""True if the tree has given filename."""
110
raise NotImplementedError()
112
def has_id(self, file_id):
113
return self.inventory.has_id(file_id)
115
__contains__ = has_id
117
def has_or_had_id(self, file_id):
118
if file_id == self.inventory.root.file_id:
120
return self.inventory.has_id(file_id)
123
return iter(self.inventory)
125
def id2path(self, file_id):
126
return self.inventory.id2path(file_id)
128
def is_control_filename(self, filename):
129
"""True if filename is the name of a control file in this tree.
131
:param filename: A filename within the tree. This is a relative path
132
from the root of this tree.
134
This is true IF and ONLY IF the filename is part of the meta data
135
that bzr controls in this tree. I.E. a random .bzr directory placed
136
on disk will not be a control file for this tree.
138
return self.bzrdir.is_control_filename(filename)
140
def iter_entries_by_dir(self):
141
"""Walk the tree in 'by_dir' order.
143
This will yield each entry in the tree as a (path, entry) tuple. The
144
order that they are yielded is: the contents of a directory are
145
preceeded by the parent of a directory, and all the contents of a
146
directory are grouped together.
148
return self.inventory.iter_entries_by_dir()
150
def kind(self, file_id):
151
raise NotImplementedError("subclasses must implement kind")
153
def _comparison_data(self, entry, path):
154
raise NotImplementedError(self._comparison_data)
156
def _file_size(entry, stat_value):
157
raise NotImplementedError(self._file_size)
159
def _get_inventory(self):
160
return self._inventory
162
def get_file(self, file_id):
163
"""Return a file object for the file file_id in the tree."""
164
raise NotImplementedError(self.get_file)
166
def get_file_by_path(self, path):
167
return self.get_file(self._inventory.path2id(path))
169
inventory = property(_get_inventory,
170
doc="Inventory of this Tree")
172
def _check_retrieved(self, ie, f):
175
fp = fingerprint_file(f)
178
if ie.text_size is not None:
179
if ie.text_size != fp['size']:
180
raise BzrError("mismatched size for file %r in %r" % (ie.file_id, self._store),
181
["inventory expects %d bytes" % ie.text_size,
182
"file is actually %d bytes" % fp['size'],
183
"store is probably damaged/corrupt"])
185
if ie.text_sha1 != fp['sha1']:
186
raise BzrError("wrong SHA-1 for file %r in %r" % (ie.file_id, self._store),
187
["inventory expects %s" % ie.text_sha1,
188
"file is actually %s" % fp['sha1'],
189
"store is probably damaged/corrupt"])
191
def path2id(self, path):
192
"""Return the id for path in this tree."""
193
return self._inventory.path2id(path)
195
def print_file(self, file_id):
196
"""Print file with id `file_id` to stdout."""
198
sys.stdout.write(self.get_file_text(file_id))
204
"""What files are present in this tree and unknown.
206
:return: an iterator over the unknown files.
213
def filter_unversioned_files(self, paths):
214
"""Filter out paths that are not versioned.
216
:return: set of paths.
218
# NB: we specifically *don't* call self.has_filename, because for
219
# WorkingTrees that can indicate files that exist on disk but that
221
pred = self.inventory.has_filename
222
return set((p for p in paths if not pred(p)))
226
from bzrlib.revisiontree import RevisionTree
229
class EmptyTree(Tree):
232
self._inventory = Inventory()
233
warn('EmptyTree is deprecated as of bzr 0.9 please use '
234
'repository.revision_tree instead.',
235
DeprecationWarning, stacklevel=2)
237
def get_parent_ids(self):
240
def get_symlink_target(self, file_id):
243
def has_filename(self, filename):
246
def kind(self, file_id):
247
assert self._inventory[file_id].kind == "directory"
250
def list_files(self):
253
def __contains__(self, file_id):
254
return (file_id in self._inventory)
256
def get_file_sha1(self, file_id, path=None, stat_value=None):
260
######################################################################
263
# TODO: Merge these two functions into a single one that can operate
264
# on either a whole tree or a set of files.
266
# TODO: Return the diff in order by filename, not by category or in
267
# random order. Can probably be done by lock-stepping through the
268
# filenames from both trees.
271
def file_status(filename, old_tree, new_tree):
272
"""Return single-letter status, old and new names for a file.
274
The complexity here is in deciding how to represent renames;
275
many complex cases are possible.
277
old_inv = old_tree.inventory
278
new_inv = new_tree.inventory
279
new_id = new_inv.path2id(filename)
280
old_id = old_inv.path2id(filename)
282
if not new_id and not old_id:
283
# easy: doesn't exist in either; not versioned at all
284
if new_tree.is_ignored(filename):
285
return 'I', None, None
287
return '?', None, None
289
# There is now a file of this name, great.
292
# There is no longer a file of this name, but we can describe
293
# what happened to the file that used to have
294
# this name. There are two possibilities: either it was
295
# deleted entirely, or renamed.
297
if new_inv.has_id(old_id):
298
return 'X', old_inv.id2path(old_id), new_inv.id2path(old_id)
300
return 'D', old_inv.id2path(old_id), None
302
# if the file_id is new in this revision, it is added
303
if new_id and not old_inv.has_id(new_id):
306
# if there used to be a file of this name, but that ID has now
307
# disappeared, it is deleted
308
if old_id and not new_inv.has_id(old_id):
315
def find_renames(old_inv, new_inv):
316
for file_id in old_inv:
317
if file_id not in new_inv:
319
old_name = old_inv.id2path(file_id)
320
new_name = new_inv.id2path(file_id)
321
if old_name != new_name:
322
yield (old_name, new_name)
325
def find_ids_across_trees(filenames, trees, require_versioned=True):
326
"""Find the ids corresponding to specified filenames.
328
All matches in all trees will be used, and all children of matched
329
directories will be used.
331
:param filenames: The filenames to find file_ids for
332
:param trees: The trees to find file_ids within
333
:param require_versioned: if true, all specified filenames must occur in
335
:return: a set of file ids for the specified filenames and their children.
339
specified_ids = _find_filename_ids_across_trees(filenames, trees,
341
return _find_children_across_trees(specified_ids, trees)
344
def _find_filename_ids_across_trees(filenames, trees, require_versioned):
345
"""Find the ids corresponding to specified filenames.
347
All matches in all trees will be used.
349
:param filenames: The filenames to find file_ids for
350
:param trees: The trees to find file_ids within
351
:param require_versioned: if true, all specified filenames must occur in
353
:return: a set of file ids for the specified filenames
356
interesting_ids = set()
357
for tree_path in filenames:
360
file_id = tree.inventory.path2id(tree_path)
361
if file_id is not None:
362
interesting_ids.add(file_id)
365
not_versioned.append(tree_path)
366
if len(not_versioned) > 0 and require_versioned:
367
raise errors.PathsNotVersionedError(not_versioned)
368
return interesting_ids
371
def _find_children_across_trees(specified_ids, trees):
372
"""Return a set including specified ids and their children
374
All matches in all trees will be used.
376
:param trees: The trees to find file_ids within
377
:return: a set containing all specified ids and their children
379
interesting_ids = set(specified_ids)
380
pending = interesting_ids
381
# now handle children of interesting ids
382
# we loop so that we handle all children of each id in both trees
383
while len(pending) > 0:
385
for file_id in pending:
387
if file_id not in tree:
389
entry = tree.inventory[file_id]
390
for child in getattr(entry, 'children', {}).itervalues():
391
if child.file_id not in interesting_ids:
392
new_pending.add(child.file_id)
393
interesting_ids.update(new_pending)
394
pending = new_pending
395
return interesting_ids
398
class InterTree(InterObject):
399
"""This class represents operations taking place between two Trees.
401
Its instances have methods like 'compare' and contain references to the
402
source and target trees these operations are to be carried out on.
404
clients of bzrlib should not need to use InterTree directly, rather they
405
should use the convenience methods on Tree such as 'Tree.compare()' which
406
will pass through to InterTree as appropriate.
412
def compare(self, want_unchanged=False, specific_files=None,
413
extra_trees=None, require_versioned=False):
414
"""Return the changes from source to target.
416
:return: A TreeDelta.
417
:param specific_files: An optional list of file paths to restrict the
418
comparison to. When mapping filenames to ids, all matches in all
419
trees (including optional extra_trees) are used, and all children of
420
matched directories are included.
421
:param want_unchanged: An optional boolean requesting the inclusion of
422
unchanged entries in the result.
423
:param extra_trees: An optional list of additional trees to use when
424
mapping the contents of specific_files (paths) to file_ids.
425
:param require_versioned: An optional boolean (defaults to False). When
426
supplied and True all the 'specific_files' must be versioned, or
427
a PathsNotVersionedError will be thrown.
429
# NB: show_status depends on being able to pass in non-versioned files and
430
# report them as unknown
431
trees = (self.source, self.target)
432
if extra_trees is not None:
433
trees = trees + tuple(extra_trees)
434
specific_file_ids = find_ids_across_trees(specific_files,
435
trees, require_versioned=require_versioned)
436
if specific_files and not specific_file_ids:
437
# All files are unversioned, so just return an empty delta
438
# _compare_trees would think we want a complete delta
439
return delta.TreeDelta()
440
return delta._compare_trees(self.source, self.target, want_unchanged,
443
def iter_changes(self, from_tree, to_tree, include_unchanged,
445
"""Generate an iterator of changes between trees.
448
(file_id, path, changed_content, versioned, parent, name, kind,
451
Path is relative to the to_tree. changed_content is True if the file's
452
content has changed. This includes changes to its kind, and to
455
versioned, parent, name, kind, executable are tuples of (from, to) if
456
changed. If a file is missing in a tree, its kind is None.
458
Iteration is done in parent-to-child order, relative to the to_tree.
460
def get_versioned_kind(tree, file_id):
462
return tree.kind(file_id)
463
except errors.NoSuchFile:
467
if specific_file_ids is not None:
468
specific_file_ids = set(specific_file_ids)
469
from_entries_by_dir = list(from_tree.iter_entries_by_dir())
470
from_data = dict((e.file_id, (p, e)) for p, e in from_entries_by_dir)
471
for to_path, to_entry in to_tree.iter_entries_by_dir():
472
file_id = to_entry.file_id
473
to_paths[file_id] = to_path
474
if (specific_file_ids is not None and
475
file_id not in specific_file_ids):
477
changed_content = False
478
from_path, from_entry = from_data.get(file_id, (None, None))
479
from_versioned = (from_entry is not None)
480
if from_entry is not None:
481
from_versioned = True
482
from_name = from_entry.name
483
from_parent = from_entry.parent_id
484
from_kind, from_executable, from_stat = \
485
from_tree._comparison_data(from_entry, from_path)
487
from_versioned = False
491
from_executable = None
492
versioned = (from_versioned, True)
493
to_kind, to_executable, to_stat = \
494
to_tree._comparison_data(to_entry, to_path)
495
kind = (from_kind, to_kind)
496
if kind[0] != kind[1]:
497
changed_content = True
498
elif from_kind == 'file':
499
from_size = from_tree._file_size(from_entry, from_stat)
500
to_size = to_tree._file_size(to_entry, to_stat)
501
if from_size != to_size:
502
changed_content = True
503
elif (from_tree.get_file_sha1(file_id, from_path, from_stat) !=
504
to_tree.get_file_sha1(file_id, to_path, to_stat)):
505
changed_content = True
506
elif from_kind == 'symlink':
507
if (from_tree.get_symlink_target(file_id) !=
508
to_tree.get_symlink_target(file_id)):
509
changed_content = True
510
parent = (from_parent, to_entry.parent_id)
511
name = (from_name, to_entry.name)
512
executable = (from_executable, to_executable)
513
if (changed_content is not False or versioned[0] != versioned[1]
514
or parent[0] != parent[1] or name[0] != name[1] or
515
executable[0] != executable[1] or include_unchanged):
516
yield (file_id, to_path, changed_content, versioned, parent,
517
name, kind, executable)
519
for path, from_entry in from_entries_by_dir:
520
file_id = from_entry.file_id
521
if file_id in to_paths:
523
to_path = osutils.pathjoin(to_paths[from_entry.parent_id],
525
to_paths[file_id] = to_path
526
if (specific_file_ids is not None and
527
file_id not in specific_file_ids):
529
versioned = (True, False)
530
parent = (from_entry.parent_id, None)
531
name = (from_entry.name, None)
532
from_kind, from_executable, stat_value = \
533
from_tree._comparison_data(from_entry, path)
534
kind = (from_kind, None)
535
executable = (from_executable, None)
536
changed_content = True
537
# the parent's path is necessarily known at this point.
538
yield(file_id, to_path, changed_content, versioned, parent,
539
name, kind, executable)
added
removed
bzrlib/tree.py
